Ovarian cancer is the second most common malignancy of the female genital tract and the most common cause of death from gynecological cancer in the United States. At present, only 20 percent of ovarian cancer patients are diagnosed at localized stage and the majority of patients are diagnosed at advanced stages characterized by poor prognosis. One of the most promising approaches to improved screening and identification of women at risk of ovarian cancer could be the use of serum biological markers. The first biomarker that was proposed for this purpose was an antigenic determinant CA125. However, CA125 as a single measurement detects only about half of the ovarian cancers at Stage I and it may be elevated in benign gynecological conditions. Consequently, CA125 used on a single occasion, as a solitary marker, does not have an adequate sensitivity and specificity as a screening test. We hypothesize that the new strategies involving use of multiple biomarkers can substantially improve the sensitivity and specificity of screening methods for ovarian cancer. With this application, we propose to concentrate on a combination of CA125 with novel promising biomarkers affecting different pathways of ovarian carcinogenesis, which include: lysophosphatidic acid (LPA), a phospholipid growth factor involved in mitogenesis and proliferation of ovarian epithelial cells; and soluble Fas (sFas), a marker of impaired apoptosis, which may lead to formation and persistence of ovarian stroma inclusion cysts and subsequent carcinogenesis. We propose to assess a potential value of the panel of serum biomarkers (CA125, LPA, sFAS) for early detection of ovarian cancer in a case-control study nested within three participating prospective cohorts: the New York University Women's Health Study; the Northern Sweden Health and Disease Study; and the Janus Serum Bank at Oslo University Hospital, Norway.